Bloom Filter Source Route Headers for LLN Overhead Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low Power and Lossy Networks (LLNs) face challenges due to large RPL control messages, which are exacerbated by the need to include full IPv6 addresses, leading to significant overhead in communication schemes, especially when multiple DAG parents are maintained to handle time-varying link qualities.
Innovation Solution
A data packet message with a routing header that accommodates both deterministic and probabilistic source routes, using a Bloom filter to compactly encode nodal source routes, reducing the need for explicit IPv6 addresses and minimizing overhead by interpolating or utilizing probabilistic routes based on predetermined criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full IPv6 addresses are included in RPL control messages to identify DAG parents, then routing reliability is improved, but message size increases significantly creating overhead
Solution Approach 1:
The patent uses Bloom filters as a compact probabilistic data structure that copies the essential routing information in a compressed form. Instead of including full IPv6 addresses in DAO messages, the Bloom filter encodes the set of DAG parent identifiers in a space-efficient manner, allowing routers to verify parent relationships without transmitting complete address information, thus reducing message size while maintaining routing reliability
Solution Approach 2:
The patent changes the parameter representation from full IPv6 addresses (16 bytes each) to Bloom filter bit arrays (configurable size, typically much smaller). This parameter transformation allows the same routing functionality to be achieved with significantly reduced message overhead, as the Bloom filter can represent multiple parent identifiers in a compact binary form that scales efficiently with the number of parents
2Reliability
If multiple DAG parents are maintained to handle time-varying link qualities, then routing robustness is improved, but the number of parents directly increases DAO message size
Solution Approach 1:
The Bloom filter efficiently copies and represents multiple parent identifiers in a compressed format. Each parent identifier is hashed into the Bloom filter structure, allowing the representation of many parents using minimal space. This enables LLN routers to maintain multiple DAG parents for robustness without proportionally increasing DAO message size, as the Bloom filter scales sub-linearly with the number of parents
Solution Approach 2:
The patent transforms the parameter set from a list of individual parent addresses to a single Bloom filter data structure. This parameter change allows the system to maintain multiple DAG parents (improving robustness against link failures) while the message size increases much more slowly than linearly, since the Bloom filter size depends on the desired false positive rate and the number of elements, not directly on the count of parents in a linear fashion
Data Source
AI summary
In one embodiment, a data packet message is provided which includes a routing header configured to accommodate both a deterministic source route and a probabilistic source route for encoding a nodal source route. The nodal source route is selectively encoded with one or both of a deterministic source route and a probabilistic source route based upon one or more predetermined criteria.